Assistive call center interface

• US 20040249636A1
• Filed: 06/04/2003
• Published: 12/09/2004
• Est. Priority Date: 06/04/2003
• Status: Active Grant

• Current Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Original Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Inventors: Junqua, Jean-Claude, Applebaum, Ted
• Granted Patent: US 7,103,553 B2
• Time in Patent Office: Days
• US Class Current: 704/231
• CPC Class Codes: G10L 15/1822 Parsing for meaning understanding

Abstract

Unstructured voice information from an incoming caller is processed by automatic speech recognition and semantic categorization system to convert the information into structured data that may then be used to access one or more databases to retrieve associated supplemental data. The structured data and associated supplemental data are then made available through a presentation system that provides information to the call center agent and, optionally, to the incoming caller. The system thus allows a call center information processing system to handle unstructured voice input for use by the live agent in handling the incoming call and for storage and retrieval at a later time. The semantic analysis system may be implemented by a global parser or by an information retrieval technique, such as latent semantic analysis. Co-occurrence of keywords may be used to associate prior calls with an incoming call to assist in understanding the purpose of the incoming call.

Claims

1. A call center information processing system capable of processing unstructured voice input, comprising:

an input port adapted for coupling to a call center telephone switch through which unstructured voice input from an incoming caller is received;

an automatic speech recognition system receptive of said unstructured voice input and operative to convert said unstructured voice input into unstructured text data;

a semantic categorization system receptive of said unstructured text data and operative to convert said unstructured text data into structured data based on features;

a data management system receptive of said structured data and operative to access at least one database of supplemental data and to form associations between said structured data and said supplemental data; and

a presentation system communicating with said data management system for providing information to a call center agent based on at least one of said structured data and said supplemental data.

2. The system of claim 1 wherein said presentation system is coupled to said call center telephone switch for providing said information to said call center agent.

3. The system of claim 1 further comprising dialog engine coupled to said data management system and operative to prompt said incoming caller to supply additional voice input.

4. The system of claim 3 wherein said dialog engine generates prompts based on at least one of said structured data and said supplemental data.

5. The system of claim 1 wherein said data management system is further operative to store said structured data as prior call records in a call history database.

6. The system of claim 5 wherein said data management system further includes a semantic analysis system to learn new features.

7. The system of claim 6 wherein said semantic analysis system employs latent semantic analysis.

8. The system of claim 1 wherein said semantic categorization system employs a parsing system.

9. The system of claim 8 wherein said parsing system employs global parser to classify a call according to semantic categories consisting of sets of keywords.

10. The system of claim 1 wherein said semantic categorization system employs an information retrieval model selected from the group consisting of Boolean retrieval models, vector retrieval models and probabilistic retrieval models.

11. The system of claim 1 wherein said data management system further includes a similar call finder system employing a semantic distance measure.

12. The system of claim 11 wherein said semantic distance measure assesses co-occurrence frequency of keywords.

13. The system of claim 1 wherein said data management system is further operative to communicate with said telephone switch to effect transfer of said incoming caller to another party through said switch.

14. The system of claim 13 wherein said data management system passes context information through said telephone switch whereby said structured data and said supplemental data is made accessible by said other party as an effect of said transfer.

15. The system of claim 1 wherein said data management system includes an input port receptive of instructions from said call center agent to alter said structured data based on knowledge of said call center agent.

16. The system of claim 1 wherein said data management system includes an input port receptive of instructions from said call center agent to alter said formed associations between said structured data and said supplemental data based on knowledge of said call center agent.

17. The system of claim 1 wherein said data management system includes an input port receptive of instructions from said call center agent to form further associations between said structured data and said supplemental data based on knowledge of said call center agent.

18. The system of claim 1 wherein said presentation system supplies other data in addition to said structured data and said supplemental data.

19. The system of claim 18 wherein said other data are selected from the group consisting of relational data, tabular data, audio/video data, and graphical data.

20. The system of claim 1 wherein said presentation system supplies a text transcript of said unstructured voice input.

21. The system of claim 1 wherein said presentation system supplies a text transcript of said unstructured voice input having selected words highlighted based on said structured data.

22. The system of claim 1 wherein said data management system includes a stored grammar that associates semantic meaning to keywords or phrases that is used to formulate queries to the incoming caller to solicit additional voice input.

23. The system of claim 22 wherein said stored grammar is a case frame grammar.

24. The system of claim 1 wherein said presentation system provides information to plural call center agents based on at least one of said structured data and said supplemental data.

25. The system of claim 1 wherein at least one of the following recited steps is performed while an incoming call from said incoming caller is in progress:

(a) converting said unstructured voice input in to unstructured text data;

(b) converting said unstructured text data into structured data; and

(c) forming associations between said structured data and said supplemental data.

26. The system of claim 1 wherein at least one of the following recited steps is performed after completion of a call from said incoming caller:

(a) converting said unstructured voice input in to unstructured text data;

(b) converting said unstructured text data into structured data; and

(c) forming associations between said structured data and said supplemental data.

27. A method of assisting call center agents based on unstructured voice input from the caller, comprising:

performing automatic speech recognition upon said unstructured voice input extracted from an incoming call to generate unstructured text data;

processing said unstructured text data by semantic categorization of said unstructured text data to generate structured data that characterizes the incoming call;

forming associations between said structured data and supplemental data obtained from a data store;

presenting information to the call center agent based on at least one of said structured data and said supplemental data.

28. The method of claim 27 wherein said processing is performed using a plurality of local parsers to identify local features within said unstructured text data.

29. The method of claim 28 wherein said processing is further performed using a global parser to identify relationships among said local features.

30. The method of claim 27 wherein said processing is performed using a predetermined set of features that are manually selected by a user.

31. The method of claim 27 wherein said processing is performed using a predetermined set of features that are determined by an automatic semantic analysis process.

32. The method of claim 31 wherein said automatic semantic analysis process employs an information retrieval model selected from the group consisting of Boolean retrieval models, vector retrieval models and probabilistic retrieval models.

33. The method of claim 27 wherein said processing step is performed using an information retrieval model selected from the group consisting of Boolean retrieval models, vector retrieval models and probabilistic retrieval models.

34. The method of claim 27 further comprising accessing a database to obtain additional information for presentation based on said supplemental data.

35. The method of claim 34 wherein said additional information is selected from the group consisting of relational data, tabular data, audio/video data, and graphical data.

36. The method of claim 27 further comprising using at least one of said structured data and said supplemental data to identify semantically similar prior calls and to extract information about said similar prior calls from a prior call history database.

37. The method of claim 27 further comprising automatically generating a query to solicit additional voice input from the caller based on said unstructured text data.

38. The method of claim 27 further comprising transferring at least a portion of at least one of said structured data and said supplemental data to a second call center agent.

39. A method of transferring knowledge between call center agents in handling an incoming call, comprising:

performing automatic speech recognition upon said unstructured voice input extracted from an incoming call to generate unstructured text data;

processing said unstructured text data by semantic categorization of said unstructured text data to generate structured data that characterizes the incoming call;

forming associations between said structured data and supplemental data obtained from a data store to define associated data; and

storing said associated data where it may be accessed by plural call agents in the handling of said incoming call.

40. The method of claim 39 wherein said processing step is performed using a parsing system.

41. The method of claim 39 wherein said processing step is performed using an information retrieval model selected from the group consisting of Boolean retrieval models, vector retrieval models and probabilistic retrieval models.